This is a joker project to test Face Recognition library. It has the following configurations/behaviours:
- Face Recognition lib using DLIB + MKL + AVX Support through a falcon endpoint;
- Python LogBook;
- Automatic logging with fluentd + docker + elasticsearch;
- Backup on S3 using Minio.
Docker and Docker-Compose installed and working.
Clone this repository and use the command docker-compose up -d
inside the cloned folder to start the server.
After that, you just need to fire the client.py
through the command:
docker-compose exec face python client.py
Just overwrite the known/base.jpg
(which is jonatahan goldsmith by default) file with the desired face.
Just put the desired images in the unknown dir.
Access the Kibana in the localhost:5601
and click on Discover Tab: the application logs will be persisted in json
format inside the elasticsearch.
TODO